Look at this photograph of the crowd at Mitt Romney's concession speech last night:

(Stephan Savoia/Associated Press)

Now look at President Obama's victory rally:

(Jonathan Newton/The Washington Post)

Rather more colorful, isn't it?

The difference in those two photographs is this election in miniature. The key conceptual mistake that the Romney campaign, the Republican Party and many pundits made in this race was assuming that the 2008 electorate was the one-off result of extraordinary dissatisfaction over George W. Bush and excitement over Barack Obama. Absent that excitement, they figured, the electorate would settle back to what Dick Morris called "normal levels," which is to say, more white people and fewer young people.

What actually happened was that the 2008 election was partly the result of excitement and anger but partly the result of long-term demographic trends benefiting the Democratic Party. And though the 2012 election didn't feature quite the same level of organic mobilization on the Democratic side, it did feature four more years of that demographic change — and that backfilled some of what was lost by a less inspired electorate. 

But the demographics don't much care who's on the ballot. In 1980, non-Hispanic whites — or what most of us would just think of as "white people" — made up 91 percent of the electorate. In 1984, that fell to 89 percent. In 1988, that fell to 85 percent. In 1992, it jumped a bit, to 88 percent. But then it fell to 83 percent in 1996, to 81 percent in 2000, to 77 percent in 2004, and to 74 percent in 2008. In 2012, it fell yet again, to 72 percent.

"This is a trend that has been going on for many years and is likely to continue for many more," says political scientist Alan Abramowitz. "The nonwhite share of the presidential electorate has increased every four years since 1992, growing by an average of 2 percentage points per election." As Jonathan Chait has noted, if the 1988 electorate had featured the same demographics as the 2008 electorate, Michael Dukakis would've been elected president.

"Obama has successfully surfed this wave," says Simon Rosenberg, president of NDN. "But he did not create it. Which is why it may be so durable for the Democrats."

This has been obvious for some time. In 2004, John Judis and Ruy Teixeira published a book called "The Emerging Democratic Majority" that outlined these trends in detail. In 2011, Teixeira and John Halpin published a paper through the Center for American Progress that ran through the data again and concluded that "a reasonable expectation for 2012 is that the minority share of voters will rise to around 28 percent." 

Sure enough, the minority share of voters in the 2012 election was 28 percent. They got it exactly right.

The bad news for Republicans isn't what happened last night. It's that it only gets worse from here. In 2016, we can expect the minority share of the electorate to rise by another two percentage points — and we can expect the economy to be in rather better shape. 

The severity of the GOP's situation has been obscured by the fact that they tend to do better in midterm elections, where the electorate is older and whiter, and by the fact that the Supreme Court, and thus the electoral college, broke their way in 2000. Then came 9/11, and they won the 2004 election convincingly. But the reality is that the Republican Party has lost the popular vote in five of the last six presidential elections, and it's only going to get harder for them from here. 

If this election proved anything, it's that the demographics are not going to backslide in the GOP's direction. Rather, the GOP will have to move in the direction of the country's changing demographics. The question of "how" will, I think, be the foremost question on the minds of the Republican Party's top strategists in the coming months.